363
GILW.4X AND GIST
[COXTRIBUTIOX
VOL.
22
FRO31 T H E C H E h l I C A L L S B O R A T O R I E S O F IOWA S T A T E COLLEGE]
Organotin Compounds Containing Water-Solubilizing Groups : Some m-Hydroxyphenyl Derivatives HENRY GILMAN A K D LE’IT’IS A, GIST, JR.
Received October 1 , 1956 Triphengl-m-hydroxyphenyltin, diphenyldi-m-hydroxyphenyltin,phenyltri-m-hydroxyphenyltin, and tetra-m-hydroxyphenyltin have been prepared. All are insoluble in water while their solubility in base increases as the number of hydroxyphenyl groups increases. Triphenyl-m-hydroxyphen yltin reacts with ethyl bromoacetate to give ethyl m-triphenylstannylphenoxyacetate. Pure azo-tin dyes could not be obtained from the reaction of triphenyl-m-hydroxyphenyltinwith p-bromo-, p-carboxy-, and p-nitrobenzenediazonium fluoborate in neutral or sodium carbonate bufferedsolutions.
A true evaluation of the merits of organotin compounds as chemotherapeutic agents has been hindered considerably by their low solubility in water. For some time, investigators in this laboratory have been concerned with increasing the water solubility of organotin compounds by introducing watersolubilizing groups into the rnolecule.’~2This report is concerned Tvith the preparation and properties of some m-hydroxyphenyl derivatives of tin. Lithiurn m-lithiophenoxide (I) mas prepared by means of a halogen-metal interconversion reaction. Subsequently, (I) was carbonated and the product was identified as m-hydroxybenzoic acid after acidification. Although no attempt has been made to determine the optimum conditions for the preparation of (I),it has been found, through carbonation studies, that a considerably better yield of this intermediate organometallic compound was obtained when n butyllithium was added to m-bromophenol than when these reagents were mixed in the reversed order. Tetra-m-hydroxyphenyltin(11) was prepared directly by means of the reaction of (I) with anhydrous tin(IT7) chloride. The following equations are illustrative of this preparation:
(11’) and phen3rltri-m-hydroxypheiiyItin (Y)were prepared, therefore, from the reaction of the Grignard reagent with the appropriate organotin chloride as indicated by the equations: (I)
+ AZgBr, --+ wz-LiOCeH4A4g13r+ LiBr
+
y(rn-LiOCGH&IgBr) ( CGH8)d-J3nCly+ (m-LiOCgHc),’Sn(CGH,)~li
+ ?I( MgBrC1)
NHaCl aqueous
(?n-LiOCsH4)ySn!CsHs)e- ____ --f ( m-HOC6H4),Sri(CBHa)aI (1111, (IV), and ( V )
The crude reaction mixtures obtained from the preparation of (11),(111),(IT),and (Ti) had a considerable amount of n-butyl bromide as well as some phenolic materials present as by-products of the reaction. The n-butyl bromide could not be removed by distillation, for, in the vicinity of 90’ extensive decomposition occurred and the resulting resin could not be purified. It was found more effective, therefore, to concentrate the hydrolyzed reaction mixture between 40 and 50“ under reduced pressure in order to avoid excessive decomposition of the desired product. The decomposition observed is attributed to the cleavage of the carbontin bond by the acidic phenols present in the crude ~ - B I C ~ H , , O+H n-C4HYLi---+w-BrC6H40Li+ n. C ~ H X reaction mixture. The cleavage of groups from m-BrCeH40Li T n-C4HYLi--+ m-LiOC6RII.i n-CIHgBr organotin compounds by acidic reagents is dependent upon the nature of the groups attached, wit,h (1) aryl groups being more readily removed than alkyl H 4Lic1 L)~~ 4(m-LioC6~f,Li)+ SnC&--+ ( T T L - L ~ O C G+ 3 , 4 Thus, although the cleavage of tetragroups, NIIlCl (aqueous) ethyltin with phenol gave an 8% yield of cleavage (m-LiOC~H~)qSn -____ +( m-HOc&)rSn product under more drastic condition^,^ a higher de(11) gree of cleavage ~vouldbe expected with the aroCleavage and redistribution reactions hare been matic organotin compounds reported here. reported to occur when organolithium compounds, The m-hydroxyphenyl derivatives of tin (11), obtained through halogen-metal interconversion re(111), (IT), and (T) were all insoluble in water. actions, are reacted with organotin halides. These Their solubility in aqueous sodium hydroxide, homundesirable side reactions do not occur if the organoever, increased as the number of hydroxy groups in lithium derivative is first converted to the correthe molecule increased. Triphenyl-m-hydroxyphensponding Grignard reagent. Triphenyl-m-hydroxypheigJltin (I IT), diphenyldi-m-hydroxyphenyltin (3) Z. 31. Rlanulkin, J . Gen. Cheni. (U.S.S.R.),16, 235
+
( I ) H. Gilniari and C. E, Arltzeri, J . Org. Chem., 15, 994
(1950). (2) H. Gilrnaii and T. C. IVu, J . Am. Chsn*~. Xoc., 77, 3228 (1955).
(1946) [Chem.Abstr., 41, 90 (1047)l; R. H. Bullard and F’. It. Holden, J . Am. Chem. Soc., 53, 3150 (1931). (4) 111. S. Kharasch and A. 1,.Flenner, J . Am. Chem. Soc., 54, 674 (1932); Meals, J . Org. Chem., 9, 211 (1944). (5j R. S a s h and G. S. Sasin, J . Org. C h e w , 20, 770 (1955).
APRIL
1957
yltin (111) was very slightly soluble in 5% aqueous sodium hydroxide; on the other hand, diphenyldim-hydroxyphenyltin (IV), phenyltri-m-hydroxyphenyltin (V), and tetra-m-hydroxyphenyltin (11) were soluble in the same medium to the extent of approximately 10, 13, and 15 parts per hundred, respectively. The preparation of triphenyl-m-hydroxyphenyltin (111) reported herein completes this series of isomeric hydroxyphenyl derivatives of tin. The other two isomers were reported earlier.'SBIt is interesting to observe that the melting points of these three isomeric compounds do not follow the same trend that has been.observed with the three isomeric dimethylaminophenyltriphenyltin comp o u n d ~ . ' ~The ~ , ~melting points of both sets of isomeric compounds have been given in Table I. Triphenyl-m-dimethylaminophenyltin is the lowest melting isomer of this series of amino derivatives, while the corresponding hydroxy derivative (111) is the highest melting isomer of the hydroxy series.
In order to explore further the scope and limitations of the formation of azo-tin dyes using p-substituted benxenediazonium fluoborates, this coupling reaction with tripheiiyl-m-hydroxyphenyltiii (111)has been investigated. Under conditions analogous to those reported' for the preparation of triphenyl-3 dimethylamino - 6 (4'- nitropheny1azo)phenyltin, (111) did not couple with either p-bromo-, p-carboxy- or p-nitrobenzenediazonium fluoborate. The starting material (111)was recovered in yields of 86, 89.5, and 92%, respectively. Since the coupling of diaBonium salts with aromatic amines and with phenols is dependent upon the availability of resonance structures (VII) and (VIII), respectively, one finds that aromatic amines couple best in the pH range 4 to 7, while the phenols couple best in the range of from 5 to 9.'O The differences observed in the behavior of (111) and triphenyl-m-dimethylaminophenyltinunder the conditions used in Run I were understandable.
-
TABLE I MELTINGPOINTSOF SOME ISOMERIC Y-CsHdSn(COH,), Melting Point "C. When "Y" is Hydroxy group Dimethylamino group dec. 176-177" 207-208" 201-203'
orlho-
nzetapara-
110-1 12O 90-9 1O 132-134"
Sodium m-triphenylstannylphenoxide [the sodium salt of (111)] reacted smoothly with ethyl bromoacetate to give a 41% yield of ethyl m-triphenylstannylphenoxyacetate (VI). Organometallic compounds of siliconlatin, and leadgcontaining an azo linkage have been prepared. The first azo-tin dye of this type to be obtained resulted from the reaction of triphenyl-p-dimethylaminophenyltin with p-nitrobenzenediazonium chloride' in an acetate buffered solution. Subsequent work7 has shown that p-substituted benzenediazonium fluoborates could be used to advantage in reactions with triphenyl-m-dimethylaminophenyltin according to the equations : lt7
+
N=N HBF, (6) C. E. Arntzen, Zowa Stale Coll. J . Sci., 18, 6 (1043) [Chem Abstr., 38,61(1944)]. ( 7 ) H. Gilman and S. D. Rosenberg, J . Am. Chnm. SOC., 74, 5580 (1952). (8) S.V. Sunthankar and H. Gilman, J . Org. Chem., 15, 1200 (1950); see also B. N. Dolgov and 0. K. Panina, J . Gen. Chem. (U.S.S.R.),18,1129(1948). (9) H. Gilman and C. G. Stuckwisch, J. Am. Chem. Soc., 64, 1007 (1942); 72,4553 (1950). .
~
.
-
VI1
COXPOUKDS
Position of Substitution
36 0
S O M E 'M-HYDROXY DERIVATIVES O F TETRAPHEh*YLTIN
VI11
An effort was made t o move into a more favorable pH range for the coupling of (111).Thus, the reaction of p-bromo-, p-carboxy-, and p-nitrobenzenediazonium fluoborate with (111) in the presence of sodium carbonate as a buffer was investigated (Run 11).In these reactions, the starting material (111) was recovered in yields ranging from 60 to 80%. In addition, however, a red oil was obtained in each coupling reaction attempted. Some tin-containing materials were present in these oils, as indicated by a qualitative test for tin.l1 Unfortunately, they could not be resolved into any pure components. The possibility has been considered that a sodium or potassium salt of (111) might be used effectively in coupling reactions with p-substituted benzenediazonium fluoborates. Investigations of this possibility are nom under way. EXPERIMENTAL
Diphenyltzn dzchloride.ls In a 1-1. flask, equipped with en air condenser, were placed 214 g. (0.50 mole) of tetraphenyltin and 130.5 g. (0.50 mole) of anhydrous tin (IV) chloride. The mixture was heated gradually to B O 0 , over a period of 1 hr., and held a t 180" for a period of 2 hr. It w-as cooled and crystallized from petroleum ether (b.p. 57-70"). After recrystallization from the same solvent, 274 g. (837,) of diphenyltin dichloride was obtained, melting at 42-44". A mixture melting point with an authentic specimen was not not depressed. Phenyltin trichlor~de.'~In a 1-1. flask fitted nith an air .-
-
-
(10) K. H. Saunders, "The Aromatic Diazo Compounds," 2nd Edition, Edward Arnold and Co , London, 1949, p. 196. (11) Oilman and Goreau, J . Org. Chem , 17, 1470 (1952). (12) All melting and boiling points are unaorrected. (13) These procedures are based on those by B. A. Kocheshkov, Ber., 62, 996 (1926), and particularly by H. Zimmer and H. Sparmann, Chenz. Ber., 87, 645 (1954). They are adapted from modifications by T. C. Wu.
3 70
CILMAN AND GIST
condenser, a thermometer, and a mechanical stirrer were placed 177 g. (0.415 mole) of tetraphenyltin and 326 g. (1.25 moles) of anhydrous tin (IV)chloride. The reaction mixture was heated with stirring a t 150" for 3 hr.; then the product was purified by distillation. Phenyltin trichloride was obtained in a 354 g. yield (ily,), boiling a t 96-97"/1.4 mm. Lithium m-lithiophenoxide (I). Run I. Eighty ml. of an et'hereal solution containing 0.08 mole of n-butyllithium was added to a cold, well stirred solution of 6.92 g. (0.04 mole) of m-bromophenol in 25 ml. of ether. The cooling bath was removed and the reaction mixture was etirred a t room temperature for 8 hr. after the addition had been completed. The product was poured jetwise into a solid carbon dioxideether slurry. The carbonated mixture was hydrolyzed after it had reached room temperature, the aqueous phase was separated, and the ethereal layer \vas extracted with 25 ml. of lOy0 sodium hydroxide. The combined aqueous phase was saturated with carbon dioxide and the phenols separated by ethereal extraction. The product obtained on acidification of the aqueous layer was recrystallized from cthanol-water to yield 2.71 g. (49%) of m-hydroxybenzoic acid, melting a t 198-199.5'. A mixture melting point with an authentic specimen showed no depression. Run 11. This run was identical with the reaction described in Run I with the single exception that the m-bromophenol was added to the n-butyllithium. mHydroxybenzoic acid was obtained in a crude yield of 2.22 g. (40%), melting over the rmge of 187-193 '. Recrystallization from ethanol-water gave 2.01 g. (36.5%) of m-hydroxybenzoic acid melting at 195-197". 2'Ftra-m,-hydroxypheny~tin (11). To a cold, well stirred solution of 52 g. (0.30 mole) of m-bromophenol in 50 ml. of ether was added 580 ml. of an ethereal solution containing 0.60 mole of n-butyllithium. The reaction mixture wae stirred a t room t,emperature overnight; then a solution containing 15.5 g. (0.058 mole) of anhydrous tin (IV) chloride dissolved in 100 ml. of bensene was added. The resulting suspension was heated at the reflux temperature for a 4 h r . petiod, and hydrolyzed by pouring it, into an ice cold solution of ammonium chloride. The organic layer waa separated and the aqueous layer was extracted with 200 ml. of ether. The organic layers were combined and dried for 24 hr. over caicium sulfa,te,The ether vas removed and t'he n-butyl bromide was distilled under reduced pressure. The black oily residue w:is crystallized lrom chloroEorni-petroleum ether (b.p. 57-70"), with difficulty, t o 1-ield 14.0 g. (48.570) of crude t,etra-m-hydroxyphenyltin (IT) melting between 170 and 182". Recrystallization from a mixture ( I : 1) of chloroform-petroleum ether (b.p. 57-70'! gave 11.0 g. of product melting over the ra,nge of 180-184" with decomposition. Hecrjrstallization from a mixt,me of ethmol and benzene gave a product whose decomposition point depended on the rate of heating. 1Yhen t,he melting point bath was heated at the rate of one degree per min. the compound decomposed between 170 and 173"; on the other hand, if the bath waa heated at the rate of 4" per min., the product melted a t 178-179" with decomposition. Further recrystallization of this product did not change t,he decomposition point. Anal. Calcd. for C24H200aSn: Sn, 24.17. Found:14 Sn, 24.12,24.22. Tri;ohenul-m-hwdroxuahenullin(111). To a cold, well stirred solution ,f 49.5 g. 70 286 mole) of m-bromophenol in 60 ml. of ether was added 830 ml. of an ethereal solution containing 0 572 mole of n-butyllithium. The reaction mixture was Ptirred a t room temperature for 12 hr., then treated with 450 ml. of an ethereal solution containing 0.572 mole of anhydrous rnaynesiiim blomide (I) One-half hr. later, 57 g (0 148 mole) of triphenyltin chloride suspended in 250 ml. (14) The organotin compounds were analyzed by decomposing a sample with concentrated sulfuric acid followed by ignition to stannic oxide according to a recently described procediire by H. Gilman and S. D. Rosenberg, J . Am. Chem. Sac., 75,3692 (1953).
VOL.
22
of ether was added. Next, the mixture was heated a t the reflux temperature for a 3-hr. period and then hydrolyzed by pouring it into an ice cold solution of ammonium chloride. The organic layer was separated and dried over anhydrous sodium sulfate. The ether was removed by distillation and the residue wae digested with 100 ml. of ethanol; then it was cooled and filtered to remove the unreacted materials. The solid obtained was crystallized from chloroform to yield 37.9 g. (55%) of crude triphenyl-m-hydroxyphenyltin (111), melting over the range of 197-203". After recrystallization from chloroform, 34.3 g. (49.5y0) of pure (111), melting a t 207-208', was obtained. Anal. Calcd. for CadHpoOSn: Sn, 26.79. Found: Sn, 26.70, 26.85. Diphenyldi-nz-hydroxyphenyltin (IV). The preparation of (IV) followed the same general procedure outlined for the preparation of (111). After 69 g. (0.40 mole) of m-bromophenol had been treated with 0.80 mole of n-butyllithium, and the mixture had been stirred a t room temperature ovcrnight,, 0.8 mole of anhydrous magnesium bromide in 435 ml. of ether was added. One-half hr. hter, 250 ml. of an ethereal solution containing 34.4 g. (0.1 mole) of diphenyltin dichloride was added. The resulting mixture was heated a t the reflux t'emperature for a period of 3 hr. and then hydrolyzed by slowly pouring it into an ice cold solution of ammonium chloride. The organic layer was separated and dried over anhydrous sodium sulfate. The ether was removed by distillation, and the residue was digested with 100 ml. of 75% methanol. The solution was cooled, the methanol removed by decantation, and the residual oil was crystallized from a mixture (1 :3) of chloroform-pet,roleum ether (b.p. 57-70') to give 19.6 g. (42.7oj,) of crude (IV), melting between 180 and 187". After t w o recrystallizations from petroleum ether (b.p. 57-70"), 16 g. (35%) of pure diphengldi-nz-hydroxyphenyltin (IV)was obtained. The melting point of this compound was somewhat, dependent, on the rate of heat'ing of the bath. When the temperature of the meking point bath was raised at a rate of 4" per min., t'he product melted a t 201-203", whereas when the temperature of the balh was ra,ised at a rate of 1" per min., the product, nielted a8t189190". An,ul. Calcd. for C:aHzoO&n: Sn, 25.85. Found: Sn, 25.82, 25.82. Phenyltri-m-hydrozyphenyl~~n (V). The preparation of (V) followed the same general procedure outlined for the ptcparation of (111).After 83 g. (0.62 mole) of nz-bromophenol had been treated with 1.24 moles of n-butyllithium, and the reaction mixture had been stirred at room temperature overnight, 1.30 moles of anhydrous magnesium bromide in 620 ml. of ether was added. One-half hr. later, 125 ml. of it benzene solution containing 30 g. (0.10 mole) cf phenyltin trichloride was added. The mixture was heated at, the reflux temperature for 4 hr., cooled, then poured slowly inLo an ice cold solution of aqueous ammonium chloride. The organic layer was separated and dried for 24 hr. over calci!im sulfate, filtered, and concentrated, first, a t atmospheric pressure, then under reduced pressure at 50". The product was then crystallized from a, (1: 1) mixture of chloroformpetroleum ether (b.p. 57-70') to yield 23.2 g. (4870) of crude (V), melting over the range 140--152'. After two recrystallizations from the same solvent pair, 19.3 g. (40.5%) of pure product was obtained, melting at 203-205'. Anal. Calcd. for ClaHzoOBSn: Sn, 24.98. Found: Sn, 24.93, 24.96. Ethyl m-triphen,yZstannylphenozZlacetate(VI).Five-tenths of a gra,m (0.022 mole) of freshly cut sodium was dissolved in 25 ml. of absolute ethanol, and the resulting solution was added slowly to a well stirred solution of 8.86 g. (0.02 mole) of triphenyl-m-hydroxyphenyltin(111)in 200 ml. of absolute ethanol. To the resulting suspension of sodium m-triphenylstannylphenoxide was added B g. (0.03 mole) of ethyl bromoacetate, and the mixture was heated at the reflux tcmperature for a period of 2 hr. It was then concentrated by distillation until crystals began to form and then drowned with
APRIL
1957
371
SOME-WL-HYDROXY DERIVATIVES O F TETRAPIIENYLTIN
TABLE I1 PREPARATIVE
DETAILSO N
THE
Benzene Diazonium Fluoborate
Grams
p-Bromop-Carboxyp-Nitro-
2.71 2.36 2.37
ATTEMPTEDCOUPLING
O F TRIPIIENYL-VL-HYDROXYPHENYLTIN I N A
Grams of
Yield of
(111)
(111)
NEUTRAL MEDIUM
Mole
Over-all Reaction Time
M.P. of
Recovered
Recovered
Recovered
0.01 0.01 0.01
4 hr. 4 hr. 3 hr.
3.80 3.98 4.09
86 % 89.57, 92 %
203-205' 202-205' 201-205"
(111)
TABLE I11 PREPARATIVE
DETAILSO N
THE
ATTEMPTED C O U P L I N G O F TRIPHENYL-m-HYDROXYPHENYLTIN BUFFERED MEDIUM
Benzene Diazonium Fluoborate
Grams
Mole
p-Bromop-Carboxyp-Nitro-
2.71 2.36 2.37
0.01 0.01 0.01
Over-all Reaction Time
Grams of Red Oil Obtained
4 hr. 4 hr.
1.83 1.61 0.91
4 hr.
I N A SODIUiU
CARBONATE
Grams of
Yield of
(111)
(111)
M.P. of (111)
Recovered
Recovered
Recovered
2.90 2.81 3.51
67 % 63 % 79 9%
201-204" 201-204° 199-204"
water. The precipitated solid was removed by filtration and hydroxyphenyltin (111) in 100 ml. of dioxane. To this was recrystallized from ethanol to give 6.5 g. (41%) of crude added 3.2 g. (0.03 mole) of sodium carbonate in 30 ml. of (VI) melting between 91 and 95'. After two recrystalliza- water. Then 0.01 mole of the p-substituted benzenediazotions from 95% ethanol, 5.1 g. (32%) of pure product was nium fluoborate dissolved in 300 ml. of a 60% solution of obtained, melting at 97-98'. 'dioxane-water was added dropwise. Two hr. after the addiAnal. Calcd. for CzsHzeOsSn:Sn, 22.43. Found: Sn, tion had been completed, the mixture waa poured into 22.62,22.56. 500 ml. of ice water and the product was separated by careReactims of triphenyLm-hydroxyphenyltin with p-substi- fully decanting off the water layer. The deep red, oily solid luted benzenediazonium JZuoborates. Run I : I n the absence of a obtained was recrystallized twice from methanol and then bu$er. In a 600-ml. beaker was placed 4.43 g. (0.01 mole) once from chloroform to obtain the recovered (111)reported of triphenyl-m-hydroxyphenyltin (111) dissolved in 60 ml. In Table 111. On concentration of the mother liquors of reof dioxane. The solution was cooled in ice and a solution crystallization, a red viscous oil waa obtained. This red oil containing 0.01 mole of the p-substituted benzenediazonium gave a qualitative test" for tin. Attempt's to isolate a pure fluoborate dissolved in 300 ml. of a 50% dioxane-water was product from this oil by crystallization ae well as by chroadded dropwise, with vigorous stirring. No color changes matography were unsuccessful. occurred. The ice bath was removed and the reaction mixture was stirred at, room temperature for 2 hr. and then Acknowledgment. This investigation was aided poured into 500 ml. of ice water. The product was removed by a Fellowship from The National Medical Fellowby filtration and purified by recrystallization from a mixture ships, Inc., through funds appropriated by the Na(1: 1) of chloroform-petroleuQ ether (b.p. 57-70"). The products obtained from the reaction of p-bromo-, p-car- tional Foundation for Infantile Paralysis. One of us boxy-, and p-nitrobenzenediazonium fluoborate with (111) (L.A.G.) wishes to express his appreciation for this have been given in Table I1 along with other pertinent data assistance. Run 11: Xodium carbonate buffered medium.In a 600-ml. beaker was placed 4,43 g. (0.01 mole) of triphenyl-mAnms, IOWA
